Real-Time Monocular Depth Estimation Using Synthetic Data with Domain Adaptation via Image Style Transfer

Monocular depth estimation using learning-based approaches has become promising in recent years. However, most monocular depth estimators either need to rely on large quantities of ground truth depth data, which is extremely expensive and difficult to obtain, or predict disparity as an intermediary...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 2800 - 2810
Main Authors Atapour-Abarghouei, Amir, Breckon, Toby P.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Adaptation models
Data models
Estimation
Neural networks
Predictive models
Training
Training data
Online AccessGet full text
ISSN1063-6919
DOI10.1109/CVPR.2018.00296

Cover

More Information
Summary:Monocular depth estimation using learning-based approaches has become promising in recent years. However, most monocular depth estimators either need to rely on large quantities of ground truth depth data, which is extremely expensive and difficult to obtain, or predict disparity as an intermediary step using a secondary supervisory signal leading to blurring and other artefacts. Training a depth estimation model using pixel-perfect synthetic data can resolve most of these issues but introduces the problem of domain bias. This is the inability to apply a model trained on synthetic data to real-world scenarios. With advances in image style transfer and its connections with domain adaptation (Maximum Mean Discrepancy), we take advantage of style transfer and adversarial training to predict pixel perfect depth from a single real-world color image based on training over a large corpus of synthetic environment data. Experimental results indicate the efficacy of our approach compared to contemporary state-of-the-art techniques.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00296